[S] Multinomial GAM

Thu, 19 Nov 1998 08:37:43 +0100

Hello S-users

I am keen to do a GAM-model of membership of 600 observations to 12
previously identified clusters based on 4 continous predictor variables.
Until now I have done this by 12 separate GAMs with family=binomial, one for
each cluster, but only because I can not figure out wether (or how) I can do
it in one single model. I would appreciate any help to this question.


Rasmus Ejrnæs
Ph.D. in Plant Ecology
Department of Landscape Ecology
National Environmental Research Institute
tel: +45 8920 1705
fax: +45 8920 1515
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news